Island division and autonomous operation method and system of active power distribution system

Abstract

The application discloses a kind of active power distribution system island division and autonomous operation method and system.The application includes: after the failure of active power distribution system, establish the island division model of active power distribution system based on loop elimination radial constraint;Establish the output model based on small hydropower station, gas power station, energy storage system, wind farm and photovoltaic power station;According to island division model and output model, construct the autonomous operation model of active power distribution system island real-time operation considering alternating current flow and steady-state security constraints;To the autonomous operation model, the active power output control of intermittent power source considering prediction error is used.The application can effectively balance the output fluctuation of intermittent power source in power distribution island, and can tolerate distributed power output prediction error to a considerable extent, and has higher solving efficiency, is conducive to realizing the online decision of the active power distribution system recovery.

Description

Technical Field

[0001] This invention relates to the field of power system technology, specifically to a method and system for islanding and autonomous operation of an active power distribution system. Background Technology

[0002] In recent years, power outages caused by natural disasters and malicious attacks have become increasingly frequent, resulting in huge economic losses and serious social impacts. For distribution systems, since the upstream power grid is often also in a fault state after these extreme events, it is difficult to directly restore power supply to the loads in the distribution system through switching operations.

[0003] The potential of distributed generation to form distribution islands and restore critical loads remains to be explored. With the widespread integration of renewable energy generation such as wind and solar power, along with energy storage and the implementation of demand-side response mechanisms, future distribution systems will evolve into controllable and adjustable active distribution systems. The integration of various distributed generation sources and controllable loads into active distribution systems is significant for promoting the consumption of intermittent renewable energy generation and reducing carbon emissions, but it also presents new challenges to the recovery of active distribution systems after faults.

[0004] In summary, how to provide an operational method for flexibly isolating active distribution system areas that cannot be restored using the main grid after extreme events is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to overcome the defects of the prior art and provide a method and system for islanding and autonomous operation of active power distribution system, which is used for operation and recovery of active power distribution system after failure.

[0006] Therefore, one technical solution adopted by the present invention is: an active power distribution system islanding and autonomous operation method, which includes:

[0007] After a fault occurs in the active power distribution system, an island partitioning model for the active power distribution system based on loop elimination of radial constraints is established.

[0008] Establish output models based on small hydropower stations, gas-fired power stations, energy storage systems, wind farms, and photovoltaic power stations;

[0009] Based on the islanding model and the power output model, an autonomous operation model for the real-time operation of an active power distribution system in islanded mode is constructed, taking into account AC power flow and steady-state security constraints.

[0010] For the aforementioned autonomous operation model, intermittent power supply active power output control that takes into account prediction errors is adopted.

[0011] Furthermore, after a fault occurs in the active power distribution system, an islanding model for the active power distribution system based on loop elimination of radial constraints is established, specifically including:

[0012] The islanding model of the active power distribution system is as follows:

[0013]

[0014]

[0015] y ij,s ≤x j,s ,

[0016] In the formula, x i,s With y ij,s , representing the states of node i and line ij in distribution island s, respectively, with 1 and 0 indicating operation and disconnection, respectively; rs is the root node of the s-th distribution island, corresponding to the node containing the s-th DG; |l| is the number of distribution lines contained in the l-th loop; x rs,s A value of 1 indicates that the s-th distribution island needs to be constructed, while 0 indicates that the s-th distribution island does not exist, and the corresponding s-th DG can be assigned to the remaining islands; N cl N is the set of nodes containing critical loads. dg S is the set of nodes where DG is located; S is the set of distribution islands, and its index element s represents the s-th distribution island; L is the set of loops contained in the distribution network, and its element l is the set of distribution lines that make up the l-th loop; N and E are the sets of nodes and lines in the distribution system, respectively.

[0017] The status of nodes and lines in the active power distribution system after islanding is represented as follows:

[0018]

[0019] In the formula, x i With y ij These represent the states of node i and line ij in the active power distribution system, with 1 and 0 indicating operation and disconnection, respectively.

[0020] Furthermore, output models are established based on small hydropower stations, gas-fired power stations, energy storage systems, wind farms, and photovoltaic power stations, specifically including:

[0021] Distributed power sources in the active power distribution system are divided into small hydropower stations and gas-fired power stations, energy storage systems, and wind farms and photovoltaic power stations. For small hydropower stations and gas-fired power stations, models are established based on their output adjustment range and speed. For energy storage systems, output models are established based on the charging and discharging power constraints and state of charge constraints that they need to meet. For wind farms and photovoltaic power stations, output models are established based on their active power output range and reactive power output range.

[0022] Furthermore, the specific content of constructing the autonomous operation model includes:

[0023] The autonomous operation model employs a rolling optimization method to progressively determine the distribution islanding operation strategy for each scheduling period. To achieve coordination of operation strategies between adjacent scheduling periods, the autonomous operation model for the t-th scheduling period is [t, t+1, ..., t+T]. l A joint autonomous operation model for the set of scheduling periods, where T l Step size optimized for rolling;

[0024] The autonomous operation model for the t-th scheduling period is described below. The autonomous operation model considers AC power flow and steady-state security constraints.

[0025] A power flow model of a distribution system in the form of a second-order cone relaxation is established based on the branch power flow model. Its expression is as follows:

[0026]

[0027]

[0028]

[0029]

[0030]

[0031]

[0032]

[0033]

[0034]

[0035] In the formula, T t = [t, t+1, ..., t+T] l [] represents the optimized set of scheduling periods, where element t represents the t-th scheduling period; and These represent the active and reactive power outputs of the DG at node i, respectively. and These represent the active and reactive power of the load removed from node i, respectively; p ij,t With q ij,t These represent the active and reactive power flows through line ij, respectively; i ij,t With v i,t These are the squares of the current amplitude at line ij and the squares of the voltage amplitude at node i, respectively. and These represent the active and reactive power demands of the load at node i, respectively; r ij With x ij These represent the resistance and reactance of line ij, respectively; M v Let v be a sufficiently large constant. To make the above constraints more compact, its value can be taken as the maximum allowable node voltage amplitude v. max The square of ; k is the node connected to node i by a path; p ki,t With q ki,t These represent the active and reactive power flows flowing through line ki, respectively; i ki,t r is the square of the line current amplitude ki; ki With x ki These are the resistance and reactance of line ki, respectively;

[0036] Considering the safety constraints of distribution islanding, including the limits that line power flow and node voltage amplitude cannot exceed, its expression is:

[0037]

[0038] In the formula: S ij The maximum apparent power allowed to flow through line ij is determined by the line's thermal stability, dynamic stability conditions, and insulation level; v min With v max These are the minimum and maximum allowable node voltage amplitudes, respectively;

[0039] The load in the active distribution system is modeled as a continuous variable, thus forming the load shedding model constraint, the expression of which is:

[0040]

[0041] In the formula: λ i,t β is the load shedding power relationship coefficient during the scheduling period t; d With β u These are the adjustment coefficients that characterize the load commissioning and shedding during adjacent scheduling periods.

[0042] Furthermore, for the aforementioned autonomous operation model, intermittent power supply active power output control that takes into account prediction errors is adopted, specifically including:

[0043] For any wind farm or photovoltaic power station g, the objective function for controlling the active power output of the intermittent power source, taking into account prediction errors, is:

[0044] In the formula, denoted as g, representing the actual active power output of the wind farm or photovoltaic power station during the dispatch period t; the superscript ^ indicates the value of the corresponding variable after solving the autonomous operation model. Let E(·) represent the active power output of DG at node g; E(·) is the penalty function, and its expression is as follows:

[0045] In the formula, e is the threshold of the penalty function E(·);

[0046] With ξ g,t Replace the objective function in the above equation And introduce the intermediate variable γ g,t The objective function for controlling the active power output of the intermittent power source, taking into account prediction errors, is represented by the following linear model:

[0047]

[0048]

[0049] Among them, the actual active power output of wind farm or photovoltaic power station g during the dispatch period t Calculate using the following formula:

[0050] In the formula, The active power of the energy storage system in the wind farm or photovoltaic power station g; For wind farms or photovoltaic power plants, predict the active power output of g; This represents the prediction error;

[0051] In the case of minimizing negative prediction error, the prediction error is negative. Taking the minimum value indicates that the actual active power output of the wind farm or photovoltaic power station is smaller than the predicted value, and the deviation is the largest. The expression for the actual active power output of the wind farm or photovoltaic power station is:

[0052] For t∈T, the prediction error is largest for positive values. At this point, the prediction error is positive, and the maximum value indicates that the actual active power output of the wind farm or photovoltaic power station is greater than the predicted value, and the deviation is the largest. The expression for the actual active power output of the wind farm or photovoltaic power station is:

[0053] If t∈T, the objective function value of active power output control under the minimum and maximum prediction error conditions obtained by solving the processing expressions for the above two cases is... and If both are 0, it means that the effects of prediction errors can be offset by adjusting the charging and discharging state and power of the energy storage system in these two worst-case scenarios. In this case, the original islanding operation strategy of the active power distribution system is feasible.

[0054] like The following constraint is added to the autonomous operation model to prevent the autonomous operation strategy from failing due to overestimation of the predicted output. This expression is used to correct for the failure of the autonomous operation strategy caused by the maximum discharge power and minimum stored energy limits of the energy storage system:

[0055]

[0056] In the formula, and Let be the energy stored by energy storage system g at the end and the beginning of the t-th scheduling period, respectively; Let g be the discharge efficiency of the energy storage system; ΔT be the actual duration of each scheduling period;

[0057] like The constraints described in the following expression are added to the autonomous operation model to avoid the failure of the autonomous operation strategy due to the low predicted output. This expression is used to correct the failure of the autonomous operation strategy caused by the maximum charging power and the maximum storage energy limit of the energy storage system.

[0058]

[0059] In the formula, The charging efficiency of the energy storage system g;

[0060] Due to the existence of prediction errors, after solving the autonomous operation model, it is necessary to verify whether the actual active power output of each wind farm and photovoltaic power station is feasible. If it is not feasible, the corresponding constraints are generated and added to the autonomous operation model, and the autonomous operation model is solved again until the obtained autonomous operation strategy is feasible when taking into account the prediction error of wind and solar power output.

[0061] Another technical solution adopted in this invention is: an active power distribution system islanding and autonomous operation system, which is used for the operation and recovery of the active power distribution system after a fault occurs, including:

[0062] Island partitioning model construction unit: After a fault occurs in the active power distribution system, an island partitioning model for the active power distribution system based on loop elimination of radial constraints is established;

[0063] Output model construction unit: Establish output models based on small hydropower stations, gas power plants, energy storage systems, wind farms, and photovoltaic power plants;

[0064] Autonomous operation model construction unit: Based on the islanding model and output model, construct an autonomous operation model for the real-time operation of the active distribution system island, considering AC power flow and steady-state security constraints;

[0065] Active power output control unit: For the autonomous operation model, intermittent power supply active power output control that takes into account prediction error is adopted.

[0066] The beneficial effects of this invention are as follows: This invention can effectively balance the output fluctuations of intermittent power sources in distribution islands, can tolerate the output prediction errors of distributed power sources to a considerable extent, and has high solution efficiency, which is conducive to realizing online decision-making for the recovery of the active power distribution system. Attached Figure Description

[0067] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0068] Figure 1 This is a flowchart illustrating the active power distribution system islanding and autonomous operation method of the present invention;

[0069] Figure 2 This is a flowchart illustrating the solution process of the autonomous operation model of this invention;

[0070] Figure 3 This is a structural block diagram of the active power distribution system islanding and autonomous operation system of the present invention. Detailed Implementation

[0071] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0072] Figure 1 This is an exemplary embodiment illustrating an active power distribution system islanding and autonomous operation method. (Reference) Figure 1 This invention provides a method for islanding and autonomous operation of an active power distribution system. This method is used for the operation and recovery of an active power distribution system after a fault occurs, including:

[0073] Step S1: After a fault occurs in the active power distribution system, establish an island partitioning model for the active power distribution system based on loop elimination of radial constraints;

[0074] Step S2: Establish output models based on small hydropower stations, gas power plants, energy storage systems, wind farms, and photovoltaic power plants;

[0075] Step S3: Based on the islanding model and the power output model, construct an autonomous operation model for real-time operation of the distribution island that considers AC power flow and steady-state security constraints.

[0076] Step S4: For the autonomous operation model, active power output control that takes into account the prediction error of intermittent renewable energy power generation is adopted.

[0077] In step S1 above, after a fault occurs in the active power distribution system, an island partitioning model for the active power distribution system based on loop elimination of radial constraints is established, specifically including:

[0078] To overcome the shortcomings of the spanning tree model, eliminate the possibility of generating loop networks, and compare the solution efficiency with the model based on virtual power flow radial constraints, an active power distribution system island partitioning model based on loop elimination radial constraints is formed.

[0079] The islanding model of the active power distribution system can be represented as follows:

[0080]

[0081]

[0082] y ij,s ≤x j,s ,

[0083] In the formula: x i,s With y ij,s ... rs,s A value of 1 indicates that the s-th distribution island needs to be constructed, while 0 indicates that the s-th distribution island does not exist, and the corresponding s-th DG can be assigned to the other islands.

[0084] After islanding, the node and line status of the active distribution system (AND) can be represented as follows:

[0085]

[0086] In the formula, x i With y ijThese represent the states of node i and line (i,j) in the active distribution system (AND), with 1 and 0 indicating operation and disconnection, respectively.

[0087] In step S2 above, the output models for small hydropower stations, gas-fired power stations, energy storage systems, wind farms, and photovoltaic power stations are established, specifically including:

[0088] Distributed power sources in active power distribution systems are categorized into small hydropower stations and gas-fired power stations, battery energy storage systems, and wind farms and photovoltaic power stations.

[0089] For small hydropower stations and gas-fired power stations, an output model is established based on their output adjustment range and speed, and its expression is as follows:

[0090]

[0091]

[0092] Where: N ht With N gt These are the sets of nodes for hydroelectric generators and gas generators, respectively. and ( and ) represent the minimum and maximum active (reactive) output of generator g, respectively; and ΔT represents the active and reactive ramp rates of generator g, respectively; ΔT is the actual duration of each scheduling period.

[0093] For battery energy storage systems, an output model is established based on the charging and discharging power constraints and state of charge constraints that must be met. Its expression is as follows:

[0094]

[0095]

[0096]

[0097] Where: N ess It is a collection of battery energy storage systems in an ADN; and ( and ) represent the active (reactive) power of the energy storage system g during discharge and charging, respectively; and These are binary variables representing the discharge and charging states of the energy storage system g, respectively. The energy stored in energy storage system g; the subscript t indicates the t-th scheduling period; and ( and ) represent the minimum and maximum active (reactive) power of the energy storage system g during charging and discharging, respectively; and These represent the charging and discharging efficiencies of the energy storage system g, respectively. and Let G be the minimum and maximum energy that the energy storage system g can store, respectively.

[0098] For wind farms and photovoltaic power stations, an output model is established based on their active power output range and reactive power output range, and its expression is as follows:

[0099]

[0100]

[0101] Where: N wf With N pv These are collections of wind farms and photovoltaic power plants in an ADN (Automatic Distribution Network). and These represent the active and reactive power of the energy storage systems in the wind farm and photovoltaic power station g, respectively. For wind farms or photovoltaic power plants, predict the active power output of g; and These represent the minimum and maximum reactive power of g for a wind farm or a photovoltaic power station, respectively.

[0102] In step S3 above, based on the islanding model and output model, an autonomous operation model for real-time operation of the distribution island, considering AC power flow and steady-state security constraints, is constructed, specifically including:

[0103] The autonomous operation model employs a rolling optimization method to progressively determine the islanding operation strategy of the distribution system for each scheduling period. To achieve coordination of operation strategies between adjacent scheduling periods, the autonomous operation model for the t-th scheduling period is [t, t+1, ..., t+T]. l A joint autonomous operation model for the set of scheduling periods, where T l Step size optimized for rolling;

[0104] The autonomous operation model for the t-th scheduling period is described below. The autonomous operation model considers AC power flow and steady-state security constraints.

[0105] A power flow model of a distribution system in the form of a second-order cone relaxation is established based on the branch power flow model. Its expression is as follows:

[0106]

[0107]

[0108]

[0109]

[0110]

[0111]

[0112]

[0113]

[0114]

[0115] In the formula: T t = [t, t+1, ..., t+T] l [] represents the optimized set of scheduling periods, where element t represents the t-th scheduling period; and These represent the active and reactive power outputs of the DG at node i, respectively. and These represent the active and reactive power of the load removed from node i, respectively; p ij,t With q ij,t These represent the active and reactive power flows through line ij, respectively; i ij,t With v i,t These are the squares of the current amplitude at line ij and the squares of the voltage amplitude at node i, respectively. and These represent the active and reactive power demands of the load at node i, respectively; r ij With x ij These represent the resistance and reactance of line ij, respectively; M v Let v be a sufficiently large constant. To make the above constraints more compact, its value can be taken as the maximum allowable node voltage amplitude v. max The square of.

[0116] Considering the safety constraints of distribution islanding, including the limits that line power flow and node voltage amplitude cannot exceed, its expression is:

[0117]

[0118] In the formula: S ij The maximum apparent power allowed to flow through line ij can be determined by the line's thermal stability, dynamic stability conditions, and insulation level; v min With v max These represent the minimum and maximum allowable node voltage amplitudes, respectively.

[0119] The load in the active distribution system is modeled as a continuous variable, thus forming the load shedding model constraint, the expression of which is:

[0120]

[0121] In the formula: λ i,t β is the load shedding power relationship coefficient during the scheduling period t; d With β u These are the adjustment coefficients that characterize the load commissioning and shedding during adjacent scheduling periods.

[0122] In step S4 above, the autonomous operation model employs active power output control that takes into account the prediction error of intermittent renewable energy generation. (Refer to...) Figure 2 The solution to the autonomous operation model can be completed, specifically including:

[0123] For any wind farm or photovoltaic power station g, the objective function for controlling the active power output of the intermittent power source is:

[0124] In the formula: The actual active power output of the wind farm or photovoltaic power station g during the dispatch period t; Let E(·) represent the active power output of DG at node g; the superscript ^ indicates the value of the corresponding variable after solving the autonomous running model; E(·) is the penalty function, and its expression is as follows:

[0125] In the formula, e is the threshold of the penalty function E(·).

[0126] With ξ g,t Replace the objective function in the above equation And introduce the intermediate variable γ g,t The objective function for controlling the active power output of intermittent power sources, taking into account prediction errors, can be represented by the following linear model:

[0127]

[0128]

[0129] The actual active power output of wind farms or photovoltaic power stations g The following formula can be used to calculate:

[0130] In the formula, The active power of the energy storage system in the wind farm or photovoltaic power station g; For wind farms or photovoltaic power plants, predict the active power output of g; This represents the prediction error. It is a random parameter whose value range is limited by the accuracy of the prediction method used and can be obtained from historical prediction data.

[0131] We will consider the feasibility of using the active power output scheduling values ​​of wind power plants and photovoltaic power plants obtained from the autonomous operation model to account for prediction errors in two cases: the case with the smallest negative prediction error and the case with the largest positive prediction error.

[0132] In the case of minimizing negative prediction error, the prediction error is negative. Taking the minimum value indicates that the actual active power output of the wind farm or photovoltaic power station is smaller than the predicted value, and the deviation is the largest. The expression for the actual output of the wind farm or photovoltaic power station is:

[0133] The prediction error is maximized when the prediction is positive. The maximum value indicates that the actual active power output of the wind farm or photovoltaic power station is greater than the predicted value, and the deviation is the largest. The expression for the actual output of the wind farm or photovoltaic power station is:

[0134] If we solve the processing expressions for the above two cases, we obtain the active power output control objective function values ​​under the minimum and maximum prediction error conditions. and If both are 0, it means that under these two worst-case scenarios, the effects of prediction errors can be offset by adjusting the charging and discharging state and power of the energy storage system. In this case, the original islanding operation strategy of the power distribution system is feasible.

[0135] like The constraints described in the following expression (41) can be added to the autonomous operation model to avoid the failure of the autonomous operation strategy due to the overestimation of the output power. This expression is used to correct the failure of the autonomous operation strategy caused by the maximum discharge power and minimum storage energy limit of the energy storage system.

[0136]

[0137] like Then the constraints described in the following expression (42) are added to the autonomous operation model to avoid the failure of the autonomous operation strategy due to the low predicted output. This expression is used to correct the failure of the autonomous operation strategy caused by the maximum charging power and the maximum storage energy limit of the energy storage system.

[0138]

[0139] Due to the existence of prediction errors, after solving the autonomous operation model, it is necessary to verify whether the actual active power output of each wind farm and photovoltaic power station is feasible. If it is not feasible, the corresponding constraints are generated and added to the autonomous operation model, and the autonomous operation model is solved again until the obtained autonomous operation strategy is feasible when taking into account the prediction error of wind and solar power output.

[0140] Example 2

[0141] Figure 3This is an active power distribution system islanding and autonomous operation system illustrated according to an exemplary embodiment. (Reference) Figure 3 This invention provides an active power distribution system for islanding and autonomous operation. This system is used for the operation and recovery of an active power distribution system after a fault occurs, including:

[0142] Island partitioning model construction unit: After a fault occurs in the active power distribution system, an island partitioning model for the active power distribution system based on loop elimination of radial constraints is established;

[0143] Output model construction unit: Establish output models based on small hydropower stations, gas power plants, energy storage systems, wind farms, and photovoltaic power plants;

[0144] Autonomous operation model construction unit: Based on the islanding model and output model, construct an autonomous operation model for real-time operation of the distribution island, considering AC power flow and steady-state security constraints;

[0145] Active power output control unit: For the autonomous operation model, intermittent power supply active power output control that takes into account prediction error is adopted.

[0146] After a fault occurs in the active power distribution system, an islanding model for the active power distribution system based on loop elimination of radial constraints is established, specifically including:

[0147] To overcome the shortcomings of the spanning tree model, eliminate the possibility of generating loop networks, and compare the solution efficiency with the model based on virtual power flow radial constraints, an island partitioning model for active power distribution systems based on loop elimination radial constraints is formed.

[0148] The islanding model of the active power distribution system is as follows:

[0149]

[0150]

[0151] y ij,s ≤x j,s ,

[0152] In the formula, x i,s With y ij,s , representing the states of node i and line ij in distribution island s, respectively, with 1 and 0 indicating operation and disconnection, respectively; rs is the root node of the s-th distribution island, corresponding to the node containing the s-th DG; |l| is the number of distribution lines contained in the l-th loop; x rs,s A value of 1 indicates that the s-th distribution island needs to be constructed, while 0 indicates that the s-th distribution island does not exist, and the corresponding s-th DG can be assigned to the other islands.

[0153] The status of nodes and lines in the active power distribution system after islanding is represented as follows:

[0154]

[0155] In the formula, x i With y ij These represent the states of node i and line ij in the active power distribution system, with 1 and 0 indicating operation and disconnection, respectively.

[0156] Establish output models based on small hydropower stations, gas-fired power stations, energy storage systems, wind farms, and photovoltaic power stations, specifically including:

[0157] Distributed power sources in the active power distribution system are divided into small hydropower stations and gas-fired power stations, battery energy storage systems, and wind farms and photovoltaic power stations. For small hydropower stations and gas-fired power stations, output models are established based on their output adjustment range and speed. For battery energy storage systems, output models are established based on the charging and discharging power constraints and state of charge constraints that they need to meet. For wind farms and photovoltaic power stations, output models are established based on their active power output range and reactive power output range.

[0158] Construct an autonomous operation model for real-time operation of distribution islands that considers AC power flow and steady-state security constraints, specifically including:

[0159] The autonomous operation model employs a rolling optimization method to progressively determine the distribution islanding operation strategy for each scheduling period. To achieve coordination of operation strategies between adjacent scheduling periods, the autonomous operation model for the t-th scheduling period is [t, t+1, ..., t+T]. l A joint autonomous operation model for the set of scheduling periods, where T l Step size optimized for rolling;

[0160] The autonomous operation model for the t-th scheduling period is described below. The autonomous operation model considers AC power flow and steady-state security constraints.

[0161] A power flow model of a distribution system in the form of a second-order cone relaxation is established based on the branch power flow model. Its expression is as follows:

[0162]

[0163]

[0164]

[0165]

[0166]

[0167]

[0168]

[0169]

[0170]

[0171] In the formula, T t = [t, t+1, ..., t+T] l [] represents the optimized set of scheduling periods, where element t represents the t-th scheduling period; and These represent the active and reactive power outputs of the DG at node i, respectively. and These represent the active and reactive power of the load removed from node i, respectively; p ij,t With q ij,t These represent the active and reactive power flows through line ij, respectively; i ij,t With v i,t These are the squares of the current amplitude at line ij and the squares of the voltage amplitude at node i, respectively. and These represent the active and reactive power demands of the load at node i, respectively; r ij With x ij These represent the resistance and reactance of line ij, respectively; M v Let v be a sufficiently large constant. To make the above constraints more compact, its value can be taken as the maximum allowable voltage amplitude v. max The square of.

[0172] Considering the safety constraints of distribution islanding, including the limits that line power flow and node voltage amplitude cannot exceed, its expression is:

[0173]

[0174] In the formula: S ij The maximum apparent power allowed to flow through line ij is determined by the line's thermal stability, dynamic stability conditions, and insulation level; v min With v max These are the minimum and maximum allowable node voltage amplitudes, respectively;

[0175] The load in the active distribution system is modeled as a continuous variable, thus forming the load shedding model constraint, the expression of which is:

[0176]

[0177] In the formula: λ i,t β is the load shedding power relationship coefficient during the scheduling period t; d With β uThese are the adjustment coefficients that characterize the load commissioning and shedding during adjacent scheduling periods.

[0178] For the aforementioned autonomous operation model, intermittent power supply active power output control, taking into account prediction errors, is adopted, specifically including:

[0179] For any wind farm or photovoltaic power station g, the objective function for controlling the active power output of the intermittent power source is:

[0180] In the formula, Let g be the actual active power output of the wind farm or photovoltaic power station during the dispatch period t; the superscript ^ indicates the value of the corresponding variable after solving the autonomous operation model; E(·) is the penalty function, and its expression is as follows:

[0181] In the formula, e is the threshold of the penalty function E(·);

[0182] With ξ g,t Replace the objective function in the above equation And introduce the intermediate variable γ g,t The objective function for controlling the active power output of the intermittent power source, taking into account prediction errors, is represented by the following linear model:

[0183]

[0184]

[0185] Among them, the actual active power output of wind farm or photovoltaic power station g The following formula can be used to calculate:

[0186] In the formula, This represents the prediction error.

[0187] In the case of minimizing negative prediction error, the prediction error is negative. Taking the minimum value indicates that the actual active power output of the wind farm or photovoltaic power station is smaller than the predicted value, and the deviation is the largest. The expression for the actual output of the wind farm or photovoltaic power station is:

[0188] For t∈T, the prediction error is the largest for positive values. At this point, the prediction error is positive, and the maximum value indicates that the actual active power output of the wind farm or photovoltaic power station is greater than the predicted value, and the deviation is the largest. The expression for the actual power output of the wind farm or photovoltaic power station is:

[0189] If t∈T, the active power output control objective function value obtained by solving the processing expressions for the above two cases is the value under the minimum and maximum prediction error conditions. and If both are 0, it means that in these two worst-case scenarios, adjusting the charging and discharging state and power of the energy storage system can offset the effects of the prediction error, and the original autonomous operation method is feasible; if The constraints described in the following expression can be added to the autonomous operation model to avoid the failure of the autonomous operation strategy due to overestimation of the predicted output. This expression is used to correct for the failure of the autonomous operation strategy caused by the maximum discharge power and minimum stored energy limitations of the energy storage system:

[0190]

[0191] like The constraints described in the following expression are added to the autonomous operation model to avoid the failure of the autonomous operation strategy due to the low predicted output. This expression is used to correct the failure of the autonomous operation strategy caused by the maximum charging power and the maximum storage energy limit of the energy storage system.

[0192]

[0193] Due to the existence of prediction errors, after solving the autonomous operation model, it is necessary to verify whether the active power output of each wind farm and photovoltaic power station is feasible. If it is not feasible, the corresponding constraints are generated and added to the autonomous operation model, and the autonomous operation model is solved again until the obtained autonomous operation strategy is feasible when taking into account the prediction error of wind and solar power output.

[0194] The present invention has been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of the invention. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and core ideas of the present invention. It should be noted that those skilled in the art can make various improvements and modifications to the present invention without departing from its principles, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

Claims

1. A method for islanding and autonomous operation of a master distribution system, characterized in that, The method comprises: After a fault occurs in the active power distribution system, an island division model of the active power distribution system based on loop elimination of radial constraints is established; An output model based on small hydropower stations, gas power stations, energy storage systems, wind farms and photovoltaic power stations is established; According to the island division model and the output model, an autonomous operation model of real-time operation of the active power distribution system island is constructed considering alternating current power flow and steady-state safety constraints; The autonomous operation model is controlled by considering the prediction error of the active output of the intermittent power source; After a fault occurs in the active power distribution system, an island division model of the active power distribution system based on loop elimination of radial constraints is established, which specifically comprises: The island division model of the active power distribution system is as follows: where x i,s and y ij,s are the state of node i and line ij in power distribution island s, respectively, and 1 and 0 represent the on and off states, respectively; rs is the root node of the s-th power distribution island, corresponding to the node where the s-th DG is located; |l| is the number of power distribution lines contained in the l-th loop; x rs,s = 1 indicates that the s-th power distribution island needs to be constructed, and 0 indicates that the s-th power distribution island does not exist, and the corresponding s-th DG can be divided into the remaining islands; N cl is the set of nodes where important loads are located; N dg is the set of nodes where DGs are located; S is the set of power distribution islands, and its index element s represents the s-th power distribution island; L is the set of loops contained in the power distribution network, and its element l is the set of power distribution lines that constitute the l-th loop; N and E are the sets of nodes and lines in the power distribution system, respectively; The node and line state of the active power distribution system after island division is represented as follows: where x i and y ij are the states of node i and line ij in the active power distribution system, respectively, with 1 and 0 representing energized and open, respectively.

2. The active power distribution system islanding and autonomous operation method of claim 1, wherein, An output model based on small hydropower stations, gas power stations, energy storage systems, wind farms and photovoltaic power stations is established, which specifically comprises: The distributed power sources in the active power distribution system are divided into small hydropower stations and gas power stations, energy storage systems, and wind farms and photovoltaic power stations; for small hydropower stations and gas power stations, a model is established according to their output adjustment range and speed; for energy storage systems, an output model is established according to the need to meet the charge and discharge power constraints and the state of charge constraints; for wind farms and photovoltaic power stations, an output model is established according to their active power output range and reactive power output range.

3. The active power distribution system islanding and autonomous operation method of claim 1, wherein, The specific content of constructing the autonomous operation model comprises: The autonomous operation model adopts a rolling optimization method to gradually determine the power distribution island operation strategy of each scheduling period. To realize the cooperation of the operation strategies of adjacent scheduling periods, the autonomous operation model of the tth scheduling period is used to determine the joint autonomous operation model of the [t, t+1, …, t+T l ] scheduling period set, where T l is the step length of the rolling optimization. The autonomous operation model of the tth scheduling period is described below, and the autonomous operation model considers alternating current power flow and steady-state safety constraints; A second-order cone relaxation form of the distribution system power flow model is established based on the branch power flow model, and the expression is as follows: where T t = [t, t+1,..., t+T l ] is the set of optimized dispatch periods, whose element t represents the tth dispatch period; and are the active and reactive power output of DG at node i, respectively; and are the active and reactive power of the load shed at node i, respectively; p ij,t and q ij,t are the active and reactive power flow through line ij; i ij,t and v i,t are the square of the current amplitude of line ij and the square of the voltage amplitude of node i, respectively; and are the active and reactive power demand of the load at node i, respectively; r ij and x ij are the resistance and reactance of line ij, respectively; M v is a sufficiently large constant, whose value is the square of the maximum allowable voltage amplitude v max of the node; k is the node connected to node i by a line; p ki,t and q ki,t are the active and reactive power flow through line ki; i ki,t is the square of the current amplitude of line ki; r ki and x ki are the resistance and reactance of line ki, respectively; Considering the safety constraints of the distribution island, including the line current and the node voltage amplitude cannot exceed the limit, and the expression is as follows: where S ij is the maximum apparent power allowed to flow through the line ij, determined by the thermal and dynamic stability conditions of the line and the insulation level; v min and v max are the minimum and maximum node voltage amplitudes allowed, respectively; The load in the active power distribution system is modeled as a continuous variable, thereby forming a load shedding model constraint, and the expression is as follows: where λ i,t is the cut-off load power relationship coefficient in the dispatch period t; β d is the load power relationship coefficient in the dispatch period t; and u are the adjustment coefficients respectively representing the load commissioning and cut-off of adjacent dispatch periods.

4. The active power distribution system islanding and autonomous operation method of claim 3, wherein, The autonomous operation model is controlled by considering the prediction error of the active output of the intermittent power source, and the specific content comprises: For any wind farm or photovoltaic power station g, the objective function of the active power output control of the intermittent power source considering the prediction error is as follows: where, is the actual active power output of the wind farm or photovoltaic power station g in the dispatch period t; the superscript is the value of the corresponding variable after solving the autonomous operation model; represents the active power output of the DG on node g; E(·) is the penalty function, whose expression is as follows: where e is a threshold value for the penalty function E(·); Replacing the objective function of the above equation by ξ g,t and introducing an intermediate variable γ g,t The objective function of the intermittent power active power control taking into account the prediction error is represented by the following linear model:​ wherein the actual active power output of the wind farm or photovoltaic plant g in the dispatch period t is calculated with the following formula: wherein, PQg is the active power of the energy storage system in the wind farm or photovoltaic plant g; PQg is the predicted active power output of the wind farm or photovoltaic plant g; is the prediction error; For the case of minimum negative prediction error, the prediction error is negative at this time, and the minimum value indicates that the actual active power output of the wind farm or photovoltaic power station is smaller than the predicted value, and the deviation is maximum, and the actual active power output expression of the wind farm or photovoltaic power station is as follows: For the maximum positive prediction error, when the prediction error is positive, the maximum value indicates that the actual active power output of the wind farm or photovoltaic power station is greater than the predicted value, and the deviation is the largest. The actual active power output expression of the wind farm or photovoltaic power station is: If the minimum and maximum prediction error cases of the active power control target function values of the above two cases are solved And Both are 0, it means that in the two worst cases, the impact of the prediction error can be offset by adjusting the charge and discharge state and power of the energy storage system, at this time the original active power distribution system island operation strategy is feasible; If The following expression describes the constraints added to the autonomous operation model to avoid overestimation of the output power leading to the failure of the autonomous operation strategy, which is used to correct the failure of the autonomous operation strategy caused by the maximum discharge power of the energy storage system and the minimum stored energy limit: wherein and Eg(t) and Eg(t+1) are the energy stored by the energy storage system g at the end and at the beginning of the t-th scheduling period, respectively; is the discharge efficiency of the energy storage system g; and ΔT is the actual length of each scheduling period. If The constraint described by the following expression is added to the autonomous operation model to avoid the failure of the autonomous operation strategy caused by the low predicted output, which is used to correct the failure of the autonomous operation strategy caused by the maximum charging power and the highest storage energy limit of the energy storage system. In the formula, is the charging efficiency of the energy storage system g; Due to the prediction error, after solving the autonomous operation model, it is necessary to verify whether the actual active power output of each wind farm and photovoltaic power station is feasible; if not, the corresponding constraint is generated and added to the autonomous operation model, and the autonomous operation model is solved again until the autonomous operation strategy obtained is feasible considering the wind and light output prediction error.

5. An active power distribution system islanding and autonomous operation system, characterized by, The system is used for the operation and recovery of the active power distribution system after a fault occurs, which comprises: An island division model construction unit: after a fault occurs in the active power distribution system, an island division model of the active power distribution system based on loop elimination of radial constraints is established; An output model construction unit: an output model based on small hydropower stations, gas power stations, energy storage systems, wind farms and photovoltaic power stations is established; An output model construction unit: an output model based on small hydropower stations, gas power stations, energy storage systems, wind farms and photovoltaic power stations is established; The autonomous operation model construction unit: according to the island division model and the output model, an autonomous operation model of the active power distribution system island real-time operation considering AC power flow and steady-state security constraints is constructed; The active power output control unit: the active power output control of the intermittent power source considering the prediction error is adopted for the autonomous operation model; The specific content of the island division model construction unit includes: The island division model of the active power distribution system is as follows: where x i,s and y ij,s are the states of node i and line ij in power distribution island s, 1 and 0 represent the on and off respectively; rs is the root node of the s-th power distribution island, which corresponds to the node where the s-th DG is located; |l| is the number of power distribution lines contained in the l-th loop; x rs,s = 1 indicates that the s-th power distribution island needs to be constructed, and 0 indicates that the s-th power distribution island does not exist, and the corresponding s-th DG can be divided into the remaining islands; N cl is the set of nodes where important loads are located; N dg is the set of nodes where DGs are located; S is the set of power distribution islands, and its index element s represents the s-th power distribution island; L is the set of loops contained in the power distribution network, and its element l is the set of power distribution lines that constitute the l-th loop; N and E are the sets of nodes and lines in the power distribution system, respectively; The node and line state of the active power distribution system after island division is represented as follows: where x i and y ij are the states of node i and line ij in the active power distribution system, respectively, with 1 and 0 representing energized and open, respectively.

6. The active power distribution system islanding and autonomous operation system of claim 5, wherein, The specific content of the output model construction unit includes: The distributed power sources in the active power distribution system are divided into small hydropower stations and gas power stations, battery energy storage systems, and wind power plants and photovoltaic power plants; for the small hydropower stations and gas power stations, the output model is established according to the output adjustment range and speed; for the battery energy storage system, the output model is established according to the need to meet the charge and discharge power constraints and the state of charge constraints; for the wind power plant and the photovoltaic power plant, the output model is established according to the active power output range and the reactive power output range.

7. The active power distribution system islanding and autonomous operation system of claim 5, wherein, The specific content of the autonomous operation model construction unit includes: The autonomous operation model adopts a rolling optimization method to gradually determine the power distribution island operation strategy of each scheduling period. To realize the cooperation of the operation strategies of adjacent scheduling periods, the autonomous operation model of the tth scheduling period is used to determine the joint autonomous operation model of the [t, t+1, …, t+T l ] scheduling period set, where T l is the step length of the rolling optimization. The autonomous operation model of the tth scheduling period is described below, and the autonomous operation model considers AC power flow and steady-state security constraints; A power distribution system power flow model in the form of a second-order cone relaxation is established based on a branch power flow model, and the expression is as follows: where T t = [t, t+1,..., t+T l ] is the set of optimized dispatch periods, whose element t represents the tth dispatch period; and are the active and reactive power output of DG at node i, respectively; and are the active and reactive power of the load shed at node i, respectively; p ij,t and q ij,t are the active and reactive power flow through line ij; i ij,t and v i,t are the square of the current amplitude of line ij and the square of the voltage amplitude of node i, respectively; and are the active and reactive demand of the load at node i, respectively; r ij and x ij are the resistance and reactance of line ij, respectively; M v is a sufficiently large constant, whose value is the square of the maximum allowable voltage amplitude v max at node i, for making the above constraints more compact; k is the node connected to node i by a line; p ki,t and q ki,t are the active and reactive power flow through line ki; i ki,t is the square of the current amplitude of line ki; r ki and x ki are the resistance and reactance of line ki, respectively; The safety constraints of the power distribution island are considered, including the line power flow and the node voltage amplitude cannot exceed the limit, and the expression is as follows: where S ij is the maximum apparent power allowed to flow through the line ij, determined by the thermal and dynamic stability conditions of the line and the insulation level; v min and v max are the minimum and maximum node voltage amplitudes allowed, respectively; The load in the active power distribution system is modeled as a continuous variable, thereby forming a load shedding model constraint, and the expression is as follows: where λ i,t is the cut-off load power relationship coefficient in the dispatch period t; β d is the load adjustment coefficient representing the adjacent dispatch period. u is the load adjustment coefficient representing the adjacent dispatch period.

8. The active power distribution system islanding and autonomous operation system of claim 5, wherein, The specific content of the active power output control unit includes: For any wind power plant or photovoltaic power plant g, the objective function of the active power output control of the intermittent power source considering the prediction error is as follows: where, is the actual active power output of the wind farm or photovoltaic power station g in the dispatch period t; the superscript is the value of the corresponding variable after solving the autonomous operation model; represents the active power output of the DG on node g; E(·) is a penalty function, the expression of which is as follows: where e is a threshold value for the penalty function E(·); Replacing the objective function of the above equation by ξ g,t and introducing an intermediate variable γ g,t The objective function of the intermittent power active power control taking into account the prediction error is represented by the following linear model:​ wherein the actual active power output of the wind farm or photovoltaic plant g in the dispatch period t is calculated with the following formula: wherein PQ, g is the active power of the energy storage system in the wind farm or photovoltaic plant g; PQ, g is the predicted active power output of the wind farm or photovoltaic plant g; is the prediction error; For the case of minimum negative prediction error, the prediction error is negative at this time, and the minimum value indicates that the actual active power output of the wind power plant or the photovoltaic power plant is smaller than the predicted value, and the deviation is maximum, and the actual active power output expression of the wind power plant or the photovoltaic power plant is as follows: For the maximum positive prediction error, when the prediction error is positive, the maximum value indicates that the actual active power output of the wind farm or photovoltaic power station is greater than the predicted value, and the deviation is the largest. The actual active power output expression of the wind farm or photovoltaic power station is: If the minimum and maximum predicted error cases of the active power control target function values of the above two cases are solved and are both 0, it means that in these two worst cases, the impact of the prediction error can be offset by adjusting the charge and discharge state and power of the energy storage system, and at this time the original active power distribution system island operation strategy is feasible. If The following expression describes the constraints added to the autonomous operation model to avoid overestimation of the output leading to the failure of the autonomous operation strategy, which is used to correct the failure of the autonomous operation strategy caused by the maximum discharge power of the energy storage system and the minimum stored energy limit: wherein and Eg(t) and Eg(t+1) are the energy stored by the energy storage system g at the end and at the beginning of the t-th scheduling period, respectively; is the discharge efficiency of the energy storage system g; and ΔT is the actual length of each scheduling period. If The constraint described by the following expression is added to the autonomous operation model to avoid the autonomous operation strategy failure caused by the low predicted output, which is used to correct the autonomous operation strategy failure caused by the maximum charging power and the highest storage energy limit of the energy storage system. In the formula, is the charging efficiency of the energy storage system g; Due to the prediction error, the actual active power output of each wind power plant and photovoltaic power plant needs to be verified after solving the autonomous operation model; if it is not feasible, the corresponding constraint is generated and added to the autonomous operation model, and the autonomous operation model is solved again until the autonomous operation strategy obtained is feasible considering the wind and light output prediction error.

Images